gauntlet-test-authoring

Write test suites from specifications without seeing the implementation.

Updated May 10, 2026
One-click install
npx skills add https://github.com/michaeljboscia/goatrodeo --skill gauntlet-test-authoring
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: gauntlet-test-authoring
Source: https://github.com/michaeljboscia/goatrodeo/tree/main/ceremonies/gauntlet
Command: npx skills add https://github.com/michaeljboscia/goatrodeo --skill gauntlet-test-authoring

SYSTEM DOCUMENTATION & REQUIREMENTS

💡 This Skill requires hypothesis, pytest, pytest-benchmark, pytest-vcr, pandas, numpy, and includes scripts (resource) and references (resource) and assets (resource) components.

What problem does it solve?

This Skill ensures that code is verified against its specifications without the verifier seeing the implementation, enforcing a strict blind testing discipline to prevent collusion and ensure thorough code testing.

Core Features & Use Cases

  • Blind Verification: Ensures that verifiers write tests based solely on specifications, not the implementation.
  • Multi-Model Skill Loading: Loads different sets of skills based on the model's role (Net, Filter, Tiebreaker) to optimize test coverage.
  • Property-Based Testing: Uses property-based testing to catch overfitting and ensure code correctness.
  • Sanity Baseline: Ensures that tests fail against a no-op stub to catch ineffective tests.
  • Mutation Testing: Integrates mutation testing to validate the strength of the test suite.
  • Output Format: Defines a structured output format for test results and metadata.
  • Use Case: In a software development project, use this Skill to verify that the code written by a developer adheres to the specified requirements without the verifier seeing the actual code.

Quick Start

Use the gauntlet-test-authoring skill to write tests for a given spec, ensuring they pass all tests without seeing the implementation.

Frequently Asked Questions about gauntlet-test-authoring

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
What is blind testing discipline and how does it prevent implementation-based test manipulation?

Blind testing discipline verifies code against specifications without the verifier seeing the implementation, preventing tests from being manipulated by implementation details and ensuring thorough code verification.

How do I write tests for a spec without seeing the implementation?

You write tests for a spec without seeing the implementation by using the gauntlet-test-authoring skill, which enforces strict blind verification to ensure tests rely solely on specification documentation.

Does this approach support property-based testing and mutation testing?

Yes, this approach applies property-based testing to catch overfitting and integrates mutation testing to validate the overall strength and correctness of the test suite.

What dependencies do I need for property-based and mutation testing in Python?

You need the hypothesis library for property-based testing and pytest for the test suite, along with pytest-benchmark and pytest-vcr to support the testing workflow.

Why do my tests pass but fail to catch code defects in unit testing?

Your tests may be ineffective or overfit. A sanity baseline fails tests against a no-op stub to catch ineffective tests, while mutation testing validates the test suite's strength to catch defects.

Can I use pandas and numpy for system testing verification?

Yes, you can use pandas and numpy within the test suite to verify code against specifications during unit, integration, and system testing in software development.